Previous studies have suggested that Epstein-Barr virus (EBV) is present in some breast cancers. Although an association of this common virus with breast cancer has implications for disease etiology and treatment, it remains controversial, largely because of methodological problems such as: 1) detected EBV was not always localized to neoplastic cells, as opposed to bystander lymphocytes; and 2) the assays were often suspect with regard to their sensitivity and specificity for EBV. Newer technologies for detecting viral pathogens (quantitative real-time polymerase chain reaction (PCR) combined with laser capture microdissection technology for localizing the virus to certain cells), now make it feasible to both quantify viral DNA in archival tissues and determine which cell tractions harbor virus. In this study, we will combine and adapt these advanced technologies to look for EBV within malignant cells of breast tumor tissues. The Specific Aims are to: 1) develop a quantitative real-time PCR assay targeting the EBV LMP1 gene by designing TaqMan probe and primers, with attention to optimizing assay sensitivity, specificity, and linearity; 2) test the utility of this LMP1 assay, as well as predeveloped quantitative PCR assays for the BamHlW and EBNA1 portions of the EBV genome, by applying them to DNA extracted from 20 model tumors (EBV-related nasopharyngeal carcinomas (NPC); 3) test the ability to localize EBV in NPC tissues by separating tumor cells from reactive cells using laser capture microdissection, and measuring viral load in each fraction using the 3 PCR assays; 4) obtain archival specimens from 100 women with invasive breast cancer, randomly sampled from a population-based cancer registry to include epidemiologically relevant characteristics already recorded in the registry, such as younger and older age, white and nonwhite race, and less and more aggressive disease (according to clinical stage, tumor size and grade, lymph node involvement, hormone-receptor status, vital status); 5) apply the 3 PCR assays to quantitate EBV in the 100 breast cancer tissues and calculate the ratio of EBV DNA to cellular (actin) DNA for each specimen; 6) use microdissection and PCR in samples with the highest EBV levels to determine if the EBV DNA is in the tumor cells or in the normal cell fraction; and, 7) conduct exploratory descriptive analyses of EBV presence in breast tumors by patient demographic and clinical variables to characterize EBV-associated cancers epidemiologically. Study strengths include consideration of an understudied, potentially treatable factor (i.e., EBV) in breast cancer, development of a novel laboratory strategy to overcome prior methodological problems in studying this association, and the first use of a carefully sampled, representative case series in such research. The new assays should facilitate, and epidemiologic findings provide direction for, larger studies of breast cancer etiology and prognosis. The work should also be relevant to possible therapeutic and preventive strategies aimed at eliminating infected cells.